CN110311860B

CN110311860B - Multilink load balancing method and device under VXLAN

Info

Publication number: CN110311860B
Application number: CN201910670362.4A
Authority: CN
Inventors: 靳彦昭; 何俭飞
Original assignee: Beijing Topsec Technology Co Ltd; Beijing Topsec Network Security Technology Co Ltd; Beijing Topsec Software Co Ltd
Current assignee: Beijing Topsec Technology Co Ltd; Beijing Topsec Network Security Technology Co Ltd; Beijing Topsec Software Co Ltd
Priority date: 2019-07-24
Filing date: 2019-07-24
Publication date: 2022-06-24
Anticipated expiration: 2039-07-24
Also published as: CN110311860A

Abstract

The invention discloses a multilink load balancing method and a device under VXLAN, wherein the method comprises the following steps: intercepting a message encapsulated by a first VXLAN tunnel terminal node VTEP 1; acquiring a pre-configured list of links to be selected; selecting a final link by adopting a specified load balancing method according to the link state obtained by analysis; modifying the source IP address and/or the destination IP address in the message based on the final link; the modified message is sent through VTEP 1. The invention can automatically judge and select the best one of the plurality of access links, shield a series of problems caused by mutual access of different operators, over-slow network response speed and insufficient utilization of bandwidth resources, and effectively avoid the problem of unreachable routing caused by network faults.

Description

Multilink load balancing method and device under VXLAN

Technical Field

The present invention relates to the field of communications, and in particular, to a method and an apparatus for balancing a multilink load in a Virtual Extensible local area network (VXLAN).

Background

In a computer network, typically, only one operating system runs on one machine, for example, a Windows system or a Linux system runs on a PC. However, as the performance of hardware is stronger, it is more common that one machine runs a plurality of systems at the same time. For example, on a Windows machine, multiple virtual machines are simultaneously enabled by VMWare software.

The above is actually a virtual machine technology, which is one of virtualization technologies. Virtualization is a technique that transforms things from one form to another. Common virtualization technologies include the above-mentioned virtual machine technology, operating system memory virtualization, network technology virtualization, and the like.

For operating system memory virtualization, when a user's application is actually running, the required memory space may be much larger than the memory size of a physical machine, and by using a memory virtualization technology, a part of a hard disk can be virtualized into a memory, which is transparent to the user.

In networking technologies, especially in data center networks with a large number of nodes, there are many virtualization technologies that have become an indispensable part of data center networks. In practice, there are many technical details on the virtualization of data center networks, but their priorities are different. At present, it can be divided into: network virtualization, network device virtualization and network function virtualization. Wherein, network virtualization: generally refers to a technique for abstracting a physical network and its components and separating network traffic from them. Such as VLAN, QinQ, VPN, tunneling, etc. Network isolation can be easily achieved by these techniques, ignoring the location and differences of the physical hardware. Network equipment virtualization: generally, a plurality of physically independent devices are integrated into a single logically-independent virtual device (multiple virtual devices), or a device is virtualized into a plurality of logically-independent virtual devices (multiple virtual devices). The former can multiply the number of ports, forwarding capability and performance specifications available to the device, and the latter can physically separate the network functions of the device into several independent units for different users. Network function virtualization: private, dedicated network devices are replaced by industry standard based servers, storage, and network devices. For example, the overlay networking technology VXLAN, etc.

The technologies are already applied in data centers, and particularly, the technology of overlay networking is the fastest to advance. The VXLAN technology is based on an IP network, adopts an encapsulation form of 'MAC in UDP', creates a large number of virtual extensible local area networks on the existing network architecture by establishing VXLAN tunnels, and tries to improve the expansion problem of large cloud computing during deployment.

VXLAN message format as shown in fig. 1, VXLAN "encapsulates" the original two-layer frame sent by a Virtual Machine (VM) as follows: VXLAN header: a VXLAN header (8 bytes) is added, which includes a VNI field of 24bits to define different tenants in the VXLAN network. In addition, the system also comprises VXLAN Flags of 8bits, which takes the value of '00001000' and two reserved fields of 24bits and 8bits respectively. Outer UDP header: the VXLAN header and the original layer two frame together serve as payload data for UDP. In the UDP header, the destination port number is fixed to 4789, and the source port number is calculated by the hash algorithm for the original two-layer frame. Outer IP head: and packaging the outer IP header. The source IP address is the IP address of the VTEP to which the source VM belongs, and the destination IP address is the IP address of the VTEP to which the destination VM belongs. An outer MAC header: and packaging the outer Ethernet header. The source MAC address is the MAC address of the VTEP to which the source VM belongs, and the destination MAC address is the MAC address of the next hop of equipment on the path reaching the destination VTEP.

VM inter-access within VXLAN network as shown in fig. 2, after ARP request reply (between VM1 and VM 2), VM1 knows the MAC address of VM2 and wants to communicate with VM 2. VTEP1 receives the packet sent by VM1, checks from the flow table of VTEP1 with the MAC address whether VM1 and VM2 belong to the same VNI, both VMs are not located in the same VNI (not being tunneled out in the same VNI), and VTEP1 already knows all address information of VM2 (MAC and VTEP 2). The VTEP1 encapsulates the new packet according to the encapsulation method described above and then passes it to the onboarding switch. The uplink switch receives the UDP data packet sent by the VTEP1, and forwards the data packet to the corresponding port after looking up the routing table. The destination VTEP, i.e., VTEP2, checks the VNI after receiving the packet, and if the VNI in the UDP packet is consistent with the VNI of VM2, decapsulates the packet and gives it to VM2 for further processing. At this point, a packet transmission is complete. The entire VXLAN-related behavior (possibly across multiple gateways) is transparent to the virtual machine, which does not feel the process of the transmission.

In the scenario of VM inter-visits within the VXLAN network mentioned above, VTEP1 selects the address of VTEP2, which is currently only one IP address (fixed when issuing the configuration). Once the IP address route is not reachable (VTEP2 down, link failure between VTEP1 and VTEP2, etc., all cause the route to be unreachable), VTEP1 will not be able to deliver packets to VTEP 2.

Disclosure of Invention

The embodiment of the invention provides a multilink load balancing method and device under VXLAN, which are used for solving the problem that once an IP address route is unreachable, a VTEP1 generated in the prior art cannot transmit a data packet to a VTEP2 under the condition of VM mutual access in the VXLAN.

The embodiment of the invention provides a multilink load balancing method under VXLAN, which comprises the following steps:

intercepting a message encapsulated by a first VXLAN tunnel terminal node VTEP 1;

acquiring a pre-configured list of links to be selected;

selecting a final link by adopting a specified load balancing method according to the link state obtained by analysis;

modifying the source IP address and/or the destination IP address in the message based on the final link;

the modified message is sent through VTEP 1.

An embodiment of the present invention further provides a device for balancing a multi-link load under VXLAN, including:

the intercepting module is used for intercepting the message encapsulated by the first VXLAN tunnel terminal node VTEP 1;

the acquisition module is used for acquiring a pre-configured link list to be selected;

the selection module is used for selecting a final link by adopting a specified load balancing method according to the link state obtained by analysis;

a modification module for modifying the source IP address and/or the destination IP address in the message based on the final link;

and the sending module is used for sending the modified message through the VTEP 1.

An embodiment of the present invention further provides a device for balancing a multilink load under VXLAN, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program when executed by the processor implementing the steps of the method for multilink load balancing under VXLAN as described above.

By adopting the embodiment of the invention, when the VTEP1 and the VTEP2 communicate, the outer layer IP header information in the VXLAN packet is modified, so that the best state of a plurality of access links is automatically judged and selected, a series of problems caused by mutual access of different operators, over-slow network response speed and insufficient utilization of bandwidth resources are shielded, and the problem of unreachable routing caused by network faults can be effectively avoided.

The foregoing description is only an overview of the technical solutions of the present invention, and the following detailed description of the present invention is provided to enable the technical means of the present invention to be more clearly understood, and to enable the above and other objects, features, and advantages of the present invention to be more clearly understood.

Drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:

fig. 1 is a schematic diagram of a VXLAN message format in the prior art;

fig. 2 is a schematic diagram of VM inter-access within a VXLAN network in the prior art;

fig. 3 is a flow chart of a method for multi-link load balancing under VXLAN in an embodiment of the invention;

fig. 4 is a detailed processing flow diagram of a method for multilink load balancing under VXLAN in an embodiment of the invention;

fig. 5 is a schematic diagram of a device for balancing multilink load under VXLAN according to a first embodiment of the present invention;

fig. 6 is a schematic diagram of a device for balancing a multi-link load under VXLAN according to a second embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

Method embodiment

According to an embodiment of the present invention, a method for balancing a multilink load under VXLAN is provided, fig. 3 is a flowchart of the method for balancing a multilink load under VXLAN according to an embodiment of the present invention, as shown in fig. 3, the method for balancing a multilink load under VXLAN according to an embodiment of the present invention specifically includes:

step 301, intercepting a message encapsulated by a first VXLAN tunnel terminal node VTEP 1;

step 302, acquiring a pre-configured list of links to be selected; further, if the link list to be selected is judged to be empty, it is determined that no replacement IP strategy is adopted, and the intercepted original message is directly sent through the VTEP 1. If the link list to be selected is not empty, step 303 is executed.

Step 303, selecting a final link by adopting a specified load balancing method according to the link state obtained by analysis; the load balancing method specifically comprises the following steps: round robin, and minimum number of connections. The link state specifically includes: link health status, and link dynamic data values, wherein the link dynamic data specifically includes: bandwidth, packet rate, and number of links.

Step 304, modifying the source IP address and/or the destination IP address in the message based on the final link;

in step 304, the method specifically includes: modifying the source IP address in the message according to the IP address of the final link; and/or modifying the destination IP address in the message according to the pre-configured destination IP address.

Step 305, sending the modified message through the VTEP 1.

Through the technical scheme, one of the plurality of access links with the best state can be automatically judged and selected.

The above technical solutions of the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

Fig. 4 is a detailed processing flow diagram of a multilink load balancing method under VXLAN according to an embodiment of the present invention, and as shown in fig. 4, the method includes acquiring a list of links to be selected, analyzing a link state according to a specified balancing method, acquiring a final link to be selected, and modifying a source IP address or a destination IP address in a packet. The method specifically comprises the following steps:

step 1: intercepting each message encapsulated by VTEP 1;

step 2: acquiring a link list to be selected;

and step 3: judging whether the list is empty or not, if so, indicating that no replacement IP strategy is adopted, and directly entering the step 7;

and 4, step 4: selecting a final link by adopting a specified load balancing method according to the link state (including the link health state) obtained by analysis; e.g., polling, minimum number of connections, etc.;

and 5: acquiring a final link to be selected;

and 6: modifying the source IP address and/or the destination IP address in the packet;

and 7: the VTEP1 submits the encapsulation message with the modified IP address upwards.

In summary, with the aid of the technical solution of the embodiment of the present invention, the VTEP1 may automatically determine and select the best access link state by modifying the outer IP header information (modifying the source address or modifying the destination address or both) in the packet, so as to shield a series of problems caused by inter-access of different operators, slow network response speed, and insufficient utilization of bandwidth resources, and meanwhile, effectively avoid the problem of unreachable routes due to network failure.

Apparatus embodiment one

According to an embodiment of the present invention, there is provided a VXLAN lower multilink load balancing apparatus, fig. 5 is a schematic diagram of a VXLAN lower multilink load balancing apparatus according to a first embodiment of the apparatus of the present invention, and as shown in fig. 5, the VXLAN lower multilink load balancing apparatus according to an embodiment of the present invention specifically includes:

the intercepting module 50 is used for intercepting the message encapsulated by the first VXLAN tunnel terminal node VTEP 1;

an obtaining module 52, configured to obtain a pre-configured list of links to be selected;

preferably, the system may further include a determining module, configured to determine whether the to-be-selected link list is empty after the obtaining module obtains the pre-configured to-be-selected link list, if the to-be-selected link list is determined to be empty, determine not to adopt a replacement IP policy, and call the sending module to directly send the intercepted original message through the VTEP1, otherwise, call the selecting module.

A selecting module 54, configured to select a final link by using a specified load balancing method according to the link state obtained through the analysis; the load balancing method specifically comprises the following steps: a polling method and a minimum connection number method; the link state specifically includes: link health status and link dynamic data values, wherein the link dynamic data specifically includes: bandwidth, packet rate, and number of links.

A modifying module 56, configured to modify the source IP address and/or the destination IP address in the message based on the final link; the modification module is specifically configured to: modifying the source IP address in the message according to the IP address of the final link; and/or modifying the destination IP address in the message according to the pre-configured destination IP address.

A sending module 58, configured to send the modified message through the VTEP 1.

In summary, with the technical solution of the embodiment of the present invention, the VTEP1 may modify the outer IP header information (modify the source address or modify the destination address or modify both) in the packet to automatically determine and select the best access link state, thereby shielding a series of problems caused by inter-access of different operators, slow network response speed, and insufficient utilization of bandwidth resources, and effectively avoiding the problem of unreachable routing caused by network failure.

Device embodiment II

An embodiment of the present invention provides a multilink load balancing apparatus under VXLAN, as shown in fig. 6, including: a memory 60, a processor 62 and a computer program stored on the memory 60 and executable on the processor 62, which computer program, when executed by the processor 62, carries out the following method steps:

Step 303, selecting a final link by adopting a specified load balancing method according to the link state obtained by analysis; the load balancing method specifically comprises the following steps: round robin, and minimum number of connections. The link state specifically includes: link health status and link dynamic data values, wherein the link dynamic data specifically includes: bandwidth, packet rate, and number of links.

Step 305, sending the modified message through the VTEP 1.

The technical scheme can automatically judge and select one of the plurality of access links with the best state.

It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A multilink load balancing method under a virtual extensible local area network (VXLAN) is applied to a tunnel terminal node (VTEP 1) and is characterized by comprising the following steps:

acquiring a pre-configured list of links to be selected;

modifying a source IP address and/or a destination IP address of an outer layer IP header in the message based on the final link, wherein the source IP address is the IP address of the VTEP to which the source VM belongs, and the destination IP address is the IP address of the VTEP to which the destination VM belongs;

sending the modified message through the VTEP 1;

the load balancing method specifically comprises the following steps: a polling method and a minimum connection number method;

the link state specifically includes: link health status and link dynamic data values, wherein the link dynamic data specifically includes: bandwidth, packet rate, and number of links.

2. The method of claim 1, wherein after obtaining the preconfigured list of links to be selected, the method further comprises:

and if the link list to be selected is judged to be empty, determining that a replacement IP strategy is not adopted, and directly sending the intercepted original message through the VTEP 1.

3. The method of claim 1, wherein modifying the source IP address and/or the destination IP address in the message based on the final link specifically comprises:

according to the IP address of the final link, modifying the source IP address in the message; and/or the presence of a gas in the gas,

and modifying the destination IP address in the message according to the pre-configured destination IP address.

4. A multi-link load balancing apparatus under a virtual extensible local area network VXLAN, the apparatus being disposed on a tunnel termination node VTEP1, comprising:

a modifying module, configured to modify a source IP address and/or a destination IP address of an outer IP header in a message based on the final link, where the source IP address is an IP address of a VTEP to which the source VM belongs, and the destination IP address is an IP address of a VTEP to which the destination VM belongs;

a sending module, configured to send the modified packet through the VTEP 1;

5. The apparatus of claim 4, wherein the apparatus further comprises: and the judging module is used for judging whether the link list to be selected is empty or not after the pre-configured link list to be selected is obtained by the obtaining module, if the link list to be selected is determined to be empty, determining that a replacement IP strategy is not adopted, and calling the sending module to directly send the intercepted original message through the VTEP1, otherwise, calling the selecting module.

6. The apparatus of claim 4, wherein the modification module is specifically configured to:

7. A multilink load balancing device under a virtual extensible local area network (VXLAN), comprising: memory, a processor and a computer program stored on the memory and executable on the processor, the computer program when executed by the processor implementing the steps of the VXLAN downlink load balancing method according to any one of claims 1 to 3.